Planners have long recommended that clients save and invest, even while they have a mortgage, since the long-term return on equities generally exceeds the interest rate on a mortgage. Yet in reality, investors don't simply choose to invest in equities because the return is higher than a fixed alternative; instead, investors demand an equity risk premium over and above the risk-free rate to make equity investing worthwhile. For the traditional investor, the equity risk premium has represented the excess return of stocks over long-term government bonds. Yet for the mortgage borrower, the available "risk-free return" isn't just a government bond, but to prepay the mortgage and eliminate the interest cost! As a result, while the investor looks for an equity risk premium over government bonds paying 2%, the mortgage borrower actually shouldn't invest in stocks unless there's an expectation to earn an equity risk premium over a mortgage interest rate that might be 4% to 5%! Consequently, clients should prepay their mortgages unless they expect a full 9%-10%+ return on equities in the current environment that sufficiently rewards them for the risk!

The inspiration for today's blog post is a recent conversation I had with another planner about the risks of keeping a mortgage while saving/investing into a portfolio, versus just paying down the mortgage. As I've written in the past, saving into a portfolio instead of prepaying a mortgage is the equivalent of buying stocks with leverage, which as we know from margin accounts can be quite risky. Yet as my fellow planner maintained, as long as stocks have an expected return that's higher than the cost of borrowing... "why not?"

The answer, I stated, is that it's not sufficient to just get a higher return from stocks than it costs to borrow on the mortgage. The borrowing cost of the mortgage - or rather, the effective return that you get by repaying the mortgage and eliminating the associated interest cost - is a guaranteed, risk-free return. Investing in stocks, on the other hand, is risky. No investor would prudently pick a highly risky return just because it's barely higher than a risk-free alternative!

In fact, as we see from the past century of investing, stocks command an 'equity risk premium' to compensate investors for the dangers of owning stocks relative to a risk-free alternative. While Treasury Bills have only returned about 3% over the past century - essentially just keeping pace with inflation and generating a real return of 0% - intermediate to long-term government bonds have generated a return of about 5% (real return of 2%), while stocks have produced a long-term return of about 10% (real return of about 7%). This means long-term investors have demanded an equity risk premium of about 5%, the excess of stock returns over longer-term-but-risk-free government bonds that can be held to maturity. (Editor's note: There is some debate about exactly how big the equity risk premium is, with some advocating amounts slightly higher or lower than 5%, but we'll use 5% relative to Treasury Bonds as a baseline.)

Notably, if it was merely sufficient for stocks to "just" outperform bonds by any amount, we wouldn't see a whopping 5% gap between the return of stocks and the return of long-term risk-free govnerment bonds over the long run. Stock prices would have been bid up to the point where they would just barely outperform bonds, and persist at that return for the indefinite future. But in reality, that's not the case. Stocks are risky, and can go through extended periods of time where they underperform bonds. Consequently, investors only buy at a price that implicitly meets their demand for a higher return - in the form of an equity risk premium over the risk-free rate - to be compensated for their investment risk.

So how does this relate to paying down a mortgage versus investing in stocks? Like the investor, the borrower would choose to invest in stocks as long as equities were anticipated to produce the desired total return - including an appropriate risk premium - over the risk-free rate. However, while the risk-free rate is the return on government bonds for the investor, the risk-free rate for the borrower is the interest rate on the mortgage that could simply be repaid instead of investing in stocks!

Which means in practical terms, the mortgage borrower shouldn't avoid pre-paying the mortgage and investing in stocks simply because equities are expected to earn a higher return. If the borrower is going to be compensated for the risk of the investment, the borrower shouldn't direct money towards stocks unless expected to earn the mortgage interest rate plus a 5% risk premium.

This situation creates an interesting dichotomy in today's environment. In a world where the 10-year bond yields about 2%, then in theory an investor earning 7% in stocks is reasonably compensated for the risk. However, only the Federal government borrows at 2%; the typical homeowner likely has a 4% to 5% mortgage interest rate. In turn, this means that it's only appropriate to direct money towards stocks if there's a reasonable expectation that stocks will earn 9% to 10% over the time horizon... even though some are forecasting a 10-year equity return as low as 4%

Of course, any particular investor will determine his/her own expected return for stocks (and in theory could select a different equity risk premium). But the fact remains that if the borrower merely expects that stocks will generate a better return than the mortgage interest rate, but not more than 5% better (or some other risk premium), the borrower is not being sufficiently compensated for the risk of the endeavor. Alternatively, as many planners have been reducing their own forecasts for the return of equities in the coming years, to the point where equities are no longer expected to earn 9%+ returns for the next decade, this suggests that planners should be directing clients to prepay/pay down mortgages for the risk-free return, as the equity returns no longer justify an investment due to the insufficient equity risk premium they offer the borrower. And this is true even if stocks are a reasonable investment relative to government bonds, because there remains a disparity between the risk-free government bond rate for the investor, and the risk-free mortgage interest rate for the borrower.

So what do you think? Is it reasonable to expect that stocks must significantly out-earn the borrowing rate in the form of an equity risk premium to justify holding onto a portfolio instead of prepaying a mortgage? Do you believe that equities will earn enough to make the investment worthwhile for a mortgage borrower? Should clients who don't feel the current outlook for stocks carries a sufficient risk premium over their loan interest rate shift to paying down the mortgage instead?

(Editor's Note: This article was featured in the Carnival of Personal Finance: The Color Wheel Edition on the blog Money Talks.)

  • Robert Henderson

    This is something I often discuss with clients. To add to the analysis, the next logical step is to anticipate what the client’s risk tolerance and asset allocation would be in retirement. This is important because many retirees become much more conservative. As such, if they opt to accrue a larger nest egg, but still maintain a mortgage, they may now actually be paying off mortgage interest with investment returns that are below their mortgage rate (and taking taxable distributions to do it).

    At minimum, I will typically coach clients to have their mortgages paid off prior to retirement, for this simple reason.

  • Mike Dayoub, CFP®

    I agree with your math let’s add a layer: risk diversification.

    The mortgage holder bears the risk of all exclusions from the HO3 (earthquake, war, nuclear, government seizure, etc.). So in a calamity insurers won’t cover, at least my client walks away with part of his portfolio elsewhere if he didn’t use his entire savings to pay down his mortgage.

    I know that’s crazy talk. Measure those “fat tails” of homeowners risk and the math will render it negligible. But avoiding the “all your eggs in one basket” approach does seem to fulfill one of the principles of risk management: limiting exposure to large losses.

    • Michael Kitces

      I’m not sure that I follow.

      In the event a calamity destroys the real estate, the reality is that the client still owes a mortgage (as many states do not treat it as a purely non-recourse), and will now have to liquidate the portfolio anyway to pay off the mortgage – assuming the portfolio has provided sufficient returns and can be liquidated quickly.

      If the client doesn’t want to have all his/her eggs in one basket, then the solution is not to allocate so much net worth to a huge residence in the first place. I fail to see how leveraging the balance sheet REDUCES risk here.

      At best it’s close in a state that has PURE non-recourse treatment of the mortgage and will not in any way harm the client’s credit rating now or in the future by walking away from the mortgage. But most states at the least will ding the credit record hard for this, if not allowing some recourse to the borrower’s other assets to pay back the mortgage.
      – Michael

      • Scott Wilson

        I think Mike is right but that his casting the scenario in all-or-nothing terms tends to obscure it.

        In fact, real estate is a market too, and the value of the asset can fluctuate just as stocks do (albeit this is usually–but not always–less volatile than equities). So, you can lose a large chunk of your investment even if the home itself is not wiped out entirely. And that can come of market fluctuation or disaster. Either way, if most of your net worth is tied up in that single asset, you have more exposure to a single event that could wipe you out.

        And if “the solution is not to allocate so much net worth to a huge residence in the first place” then what do you propose the person does with whatever the remainder of their net worth is, if not invest it?

  • Jay Schuman

    It should also be discussed what their time frame in the home is going to be. Its more than just crunching numbers, but determining if a paid off house is the best option in this housing market. Selling with the mortgage paid in full is usually a plus, but what if you are paying off another 15-20 years of a mortgage, but may move long before that time frame. You have put all of your money into a home that you won’t own for the long-term and you aren’t guaranteed you will get it all back in a sale.

    • Michael Kitces

      The value of your home, including any hit its value will take in a sale, is the same regardless of whether the mortgage has been paid off or not.

      If the client is concerned about having so much money tied up in a piece of real estate, the problem is not how the real estate is FINANCED, it’s the fact that the client BOUGHT so much real estate in the first place.

      A big asset that consumes much of the balance sheet is the same big asset regardless of whether it was financed with cash or debt.

      – Michael

      • Jay Schuman

        Mike, a house is almost always a person’s biggest investment. It usually requires the most in cash flow and usually requires funding via a mortgage; often even for a relatively inexpensive home.

        If one puts all of their free cash flow into a home to pay off a mortgage not due for 15-20 years lets say. Then they sell it in less than 5 years and in this market get much less in sale then the actual mortgage (certainly a possibility for many today), where is the advantage?

        I am not saying your thesis is not correct, I am just suggesting there is more than just numbers to be considered. It is also important to look at the individual situation with re: to time expected in home, cost of mortgage vs current value of home, etc.

        Best regards,


        • Michael Kitces

          If the client buys a house for $500k that drops in value to $400k, those numbers are in no way impacted by whether the client paid down the mortgage or not. The asset has experienced the same loss.

          And strictly speaking, the benefit of paying down the mortgage is that the client CAN sell it and move in 5 years; if the debt is too high and the mortgage is underwater, the client has to come up with outside funds to make up the difference. Which means now the client has speculated that stock investing will outperform the mortgage interest rate over an even shorter period of time of just 5 years – which is even riskier than when the mortgage is being kept for 15-20+ years!

          But the bottom line is that changes in value of the asset – the house – have no relationship to whether the purchase was FINANCED with a loan or from cash. The cost/benefit of financing versus using cash/investments is whether the investments outearn the cost of borrowing – except when you add a required risk premium, and then acknowledge the client might want to move in 5 years, the transaction is even riskier.

          Do we really advocate that clients should invest in stocks with the assured expectation that the stocks will outperform 9% (loan rate plus risk premium) over a 5-year time horizon (the time period you suggested the client might want to move)?

          – Michael

          • Tobin Woodruff

            I’d be curious to hear if anyone knows of any empirical studies looking into what impact, if any, the seller’s having a substantial mortgage has on final sale price. Rationally speaking, what Michael says makes perfect sense: the presence or absence of a mortgage shouldn’t have any impact on the sale price. But real people often do not behave rationally. My unsubstantiated hunch is that a statistically significant number sellers without a mortgage would be willing to make a larger price concession to close the deal than sellers who are staring their mortgage in the face, because sellers with a significant mortgage have a hard number in mind that they don’t want fall below or approach too closely.

  • Ted White


    This is a really interesting post! Question for the group: How would inflation impact this analysis? If you’ve locked in a 30-year fixed-rate mortgage at 4%, inflation actually becomes your friend, allowing you to pay back the loan in later years with “dollarettes”.



    • Michael Kitces

      Inflation pushes up the nominal value of your future income.

      That’s true regardless of whether you take it and save it in future years, or you take it to repay debt in future years.

      The difference is that in the latter scenario, you’re losing a real economic cost – in the form of interest – for what will be the same future inflation-adjusted income, regardless of the presence of the mortgage.

      In other words, there’s nothing about inflation that makes a mortgage better. Yes, the mortgage is easier to repay with future inflated salary dollars, but that’s a function of inflation impacting your salary, regardless of the mortgage!
      – Michael

    • JPizzle

      Ted, I’ve been thinking the same thing. I considered plowing my savings into paying down my 4.5% mortgage quickly because it is a risk free return on investment, but as I considered the macro environment and the inflationary QE speak coming out of DC, along with the impending fiscal meltdown, I also began to consider the ramifications of not paying my mortgage early (basically going long on the inflation play) and putting all savings into equities (inflation proof). If inflation skyrockets over the next 30 years, my mortgage payment in real dollars will become a much smaller amount of my inflation adjusted nominal pay. It is a tough decision, but when you look at the state of our country where the populace is voting itself largesse from the treasury (see France & Greece), there is no discipline to correct our finances and inflation will be the end game. The only thing that will curtail inflation in this country is that Europeans will flood our treasury looking for a safer currency than the Euro.

  • Tom

    I’m not a professional– just a homeowner and investor, so forgive me if I am thinking fuzzily. My concern has always been about the years *after* the loan is repaid. To take the extreme case, suppose I buy my house for cash (ie., repay the loan in zero years). Which asset, the house or a stock portfolio, will have a higher rate of return? I’ve lived in a relatively low cost market for the past 25 years, and we experienced neither a significant boom or bust with the bubble. My house has appreciated something like 100% over the 25 years. The S&P 500 is up something like 500% over that period, right?

    • Michael Kitces

      Yes, asset-for-asset, equities have a better long-term appreciation rate than real estate, especially real estate held for personal use where there is no rental income.

      Some recent research has suggested that over the long run – smoothing out the booms and busts along the way – the real (inflation-adjusted) return on real estate may be little more than 0% – 1%/year, while equities clearly generate a higher return (albeit with higher risk and a greater danger of extended periods of underperformance).

      In the extreme, that would make the case that more people should rent their homes to free up assets to invest in equities instead of their real estate. The only caveat from that end is that owning your home is effectively an inflation hedge – it hedges the risk/impact of inflation pushing up rental housing costs. But that’s a function of the asset – you own the house you live in – not whether you finance it with debt or cash.

      – Michael

  • Mike McGinley

    There are soooo many variables that go into this decision. A big one is the randomness of returns and the fact that the investor will be dollar cost averaging into a portfolio. If treasury yields are at 6% in five years and stocks are priced to return 10% then refinancing into a thirty year mortgage at 4% is probably wise. But would an investor stick with dollar cost averaging if the markets fall substantially in a short period? Too many predictions necessary. I say, is 4% guaranteed return good? If the answer is yes then pay down the mortgage.

  • Ben Birken

    Assuming that Congress doesn’t do away with the interest deduction, shouldn’t the reference point of the guaranteed return on pre-paying be the after-tax cost of the mortgage? That should be compared to the after-tax cost of an equity portfolio. But if we assuming it’s a taxable account we’re talking about (where tax drag involves preferential cap gains treatment), wouldn’t that lower the risk premium estimate?

    • Michael Kitces

      Yes, strictly speaking, it should be the after-tax portfolio return versus the after-cost of borrowing (assuming you want to extend the current tax regime indefinitely).

      That’s still only a modest adjustment, though, as the tax arbitrage isn’t THAT big. Maybe it takes a 5% risk premium down to 4.X%? Same principle still applies.
      – Michael

  • Tom DJ

    I like this strategy, and go through this risk mitigation discussion with many clients. In some cases it comes down to psychology. Some people, even with anticipated risk premium levels of 5%+, will still choose to pay down the mortgage due to their financial psyche. They need that zero on the balance sheet, and that’s just fine.

    My only concern with the strategy is having a very good cash/available investment reserve in the event of a disability or extended job layoff while you’re in the midst of paying down the mortgage. The lender does not care that you’ve made double payments every month for the last 5 years…no income=no way to refinance to get at the equity.

  • jim schwartz

    The invest and keep the mortgage exposes the inherent conflict of interest of AUM planners – regardless fee only or not.

    The question is managing goals not managing assets in terms of priority.

    And risk tolerance measures – absent where a client is relative to each goal – is worthless.

    Looking back at now 35 years involved with the business – 20 as a fee only planner (retired from practice now 16 but still writing) it amazes me some of the biggest names how they have rationalized the big mortgage and continuing to invest rather than paying down.

    And every one – every damn one of them – that’s right every – was compensated on an asset under management compensation system. Forget projecting housing markets – the planners didn’t take into account the client’s own business/profession/work beta not to mention the psychic income of having the house paid off – and if necessary later a reverse mortgage cushion.

    As such these planners were no different than the transaction oriented product pushers – commission light.

    As for housing itself- clients and planners miss the point just like they have on insurance.

    Insurance is risk transference (period) and housing is shelter- not a piggy bank. Go to R Wilson analysis of housing and basically over time it has about equalled inflation(see bloomberg for the chart).

    And all the rest are rationalizations or as the great Rabbi Hillel was, ‘all the rest is just commentary.’

  • Bruce Steiner

    There can be a tax benefit to the extent the interest on the mortgage is deductible against ordinary income and some or all of the investment income and gains are qualified dividends, long-term capital gains, tax-exempt income, or unrealized appreciation not currently taxable.

    If you want to limit your risk, you can view the mortgage as being short a bond and long a call on a bond (the right to prepay the mortgage) in deciding upon your asset allocation.

  • Geoff Wells


    First of all, excellent post. I am a daily GenY follower and greatly appreciate your contribution to the industry.

    One item I would like you to comment on is the luck factor of equities.

    Looking at two scenarios, I see a significant luck factor present:
    1) Paying off a mortgage in 15 years, then investing the same amount of money in year 16-30 into equities.
    2) Paying off a mortgage in 30 years and investing the payment difference between the 30 year mortgage and 15 year mortgage into equities for all 30 years.

    Both scenarios have the exact same Present Value with a fixed interest rate over the period of time. With equities, the longer time horizon increases the chances avoiding “bad timing” for retirement planning. In your opinion, would this luck factor be a greater risk reducer for a client than the risk-free rate delta (mortgage rate – government bonds)?

    In addition, the “option” of refinancing a mortgage to a lower fixed rate in the future also has a value that should be considered.


  • Jason Hiley

    Michael – I liked the article and it brings up an important topic that many of us deal with every day. I agree with your premise and the logic. However there is another variable I always discuss that is hard to predict: future interest rates.

    In your argument you are assuming the only other option for the use of the money to outperform the mortgage is equities. However I think we are all a little guilty of some recency bias in relation to interest rates. It was not that long ago (though to most it seems like an eternity) that I could go down to my local bank and get 5.5% in a FDIC insured CD. So when talking to clients about wheter or not to keep a mortgage I feel like you need to talk about this issue as well. If (when?) rates rise, there is a good possiblity you will be able to lock in risk free return in excess of your fixed mortgage. If that does not happen for 10 years it’s probably not a good deal, if that happens in 2 then the client will probably make out all right.

  • Anonymous

    Michael, a tip of the hat from me for laying out a logical argument that points out an inherent conflict of interest in the AUM model. Given a forecasted (or keep it simple, historic) risk-return relationship for equities whatever the planner THINKS (i.e. rationalizes) is irrelevant compared to prudent advice for the client given their risk tolerance.

  • Jim

    Michael, I am a Tax Accountant, and I see another side of this equation. If someone has the ability to write a check tomorrow for the balance of their loan, your discussion is outstanding. However, for the majority of people I see, that is not the case. I always suggest paying down a mortgage, because in the human equation, people tend to spend the money on things that do not help them financially (car, boat, frivolous vacations, clothing, etc.). This leaves them with less money and no equity in their home.

    I have never met anyone who regrets paying off a mortgage. The look of relief when that weight is no longer there is absolutely the best look and feeling I have ever seen. If the world goes to hell in a hand basket, they just lock their door. So no matter how you look at the numbers, there really is no quantifying that feeling.

    • Michael Kitces

      Thanks for the comment.

      I would characterize this the same way even if a lump sum is not available, though. Each marginal dollar that goes into a portfolio as ‘savings’ instead of prepaying the mortgage is subject to the same framework on a mini-basis.

    • Holly P. Thomas

      Bravo, Jim and Michael. My experience is any lumpsum is good but complete freedom from debt pays gains worth more than just the interest. It adds quality years to clients’ lives.

  • Jedi

    One thing I’d like to mention is that when you are prepaying your mortgage, your monthly payment going toward interest and principal does not change until the principal is paid off. This is generally because your amortization schedule is not updated. If something occurs where perhaps early in the loan, you made prepayments but are not able to keep up, you are not ahead in your mortgage until the end. Consider saving the money with the intention of a lump sum prepayment. During the recession, some homeowners had made prepayments, but after losing their jobs and being underwater, lost their home. If they had saved their money, they would at least have been able to use that money to start over.

    • Michael Kitces

      The payment schedule may not change, but that doesn’t alter the fact that you ARE saving interest by prepaying principle on the mortgage. After all, the mortgage doesn’t last for 30 years when you prepay the mortgage; it’s just that you don’t reset your payment commitment (until it’s refinanced or paid off).

      Keeping money invested on the side with the goal of prepaying in a lump sum in the future involves taking investment risk over a relatively short time horizon (your target lump sum prepayment period) and can actually be much riskier.

      Of course, prepaying a mortgage doesn’t change the importance of having a legitimate emergency fund, to handle shorter-term liquidity needs.

      – Michael

      • Holly P. Thomas

        Jedi and Michael
        Just FYI, with many mortgage companies, if you make a large enough prepayment, you can request a reamortization which will lower the monthly payment. Not saying this is always advisable, but thought it is worth knowing about.

  • Anonymous

    Stocks are risky, but margin accounts are not inherently risky. You will not get a margin call from not paying off your mortgage.

    Also, one always has the option to pay off their mortgage. So if you have not received the return on investments necessary to pay off your mortgage, cannot you just wait until you do? At the point in time when your investments have returned more than your mortgage interest rate, cash in and pay off your mortgage.

    • Michael Kitces

      Certainly, the fact that margin loans can be called and mortgages generally cannot makes the mortgage a more appealing way to GET leverage, but it doesn’t change the financial ramifications of the deal or the impact of failing to earn a proper risk premium for the investment.

      It’s also worth noting that mortgages are generally self-amortizing and do require ongoing cash flows to support. If you want to get an interest-only or negative amortization loan to reduce the cash flow requirements, that’s certainly a potential option, but the riskiness of those loans has shown itself in the fact that they are rarely even available anymore because they so often turned out poorly. And again, regardless of the particular structure of the loan, it doesn’t change the fact that it’s investing with leverage, and has the possibility of generating significant losses or failing to earn an appropriate risk premium.

      – Michael

  • Pingback:

  • Eric

    Exactly. Would you recommend that your client take out a 4 to 5% personal loan to invest in the market? Than why would recommend they keep their mortgage to invest more in the market. It’s the same thing. If you pay off the debt, you are GUARENTEED to get the return equal to your interest rate. That is hard to beat if it’s a 4 to 5% risk free rate of return.

  • Pingback:

  • Anonymous


    Really interesting posts (both this one and the one it links to). Isn’t the logical extension of this argument that one shouldn’t buy a home with a mortgage in the first place, and should just rent and invest in a diversied portfolio instead? I’ve been thinking about this a lot as a millennial approaching the typical home buying age, and cannot come up with a rational argument for one buying unless the rent vs buy math is overwhelmingly in favor of renting, which it isn’t in hot markets like the Bay Area. What’s your perspective on this.


2 Tweets / Trackbacks

Michael E. Kitces

I write about financial planning strategies and practice management ideas, and have created several businesses to help people implement them.

For ConsumersFor Advisors

Blog Updates by Email

Nerd’s Eye View Praise

@MichaelKitces Twitter

Out and About

Wednesday, September 2nd, 2015

*Understanding Longevity Annuities and their Potential Role in Retirement *Trends & Developments in Long-Term Care Insurance *Understanding the New World of Health Insurance @ FPA Illinois

Wednesday, September 9th, 2015

*Future of Financial Planning in the Digital Age *Modern Portfolio Theory 2.0 @ FPA San Diego

Thursday, September 17th, 2015

*Future of Financial Planning in the Digital Age *Social Media for Financial Planners *Understanding Longevity Annuities and their Potential Role in Retirement Income @ FPA Colorado